L-Glutamate as a Novel Modifier of Root Growth and Branching

Abstract

Exogenous L-glutamate (L-Glu) has been shown to be able to elicit major changes in Arabidopsis root architecture at micromolar concentrations. The root response, which is strongly genotype-dependent, is specific to L-Glu and involves both inhibition of primary root growth and stimulation of root branching behind the primary root tip. The L-Glu appears to be sensed directly at the root tip, where it inhibits meristematic activity. An intriguing and still unanswered question is whether members of the family of Glu receptor-like genes (GLRs) have a role in mediating this response. A pharmacological approach described here, using agonists and antagonists of mammalian ionotropic Glu receptors, has failed to resolve the issue. Progress towards identifying the genes involved in the root response to L-Glu is likely to come through the application of forward and reverse genetics, in combination with quantitative trait loci (QTL) mapping.

Exogenous L-Glu is Able to Elicit Marked Changes in Root Architecture

Many plant species are able to proliferate their roots preferentially within nutrient-rich soil patches.^1,2 This type of response can be seen as a form of foraging behaviour that serves to enhance the plant's ability to compete for limiting supplies of nutrients.³ Until recently, only inorganic nutrients (such as NO₃⁻ and Pi) were known to be capable of eliciting this response, but a recent report has shown that an amino acid (L-Glu) can trigger changes in root growth and branching that could potentially constitute a novel type of adaptive response.⁴

Under aseptic conditions, L-Glu applied exogenously to Arabidopsis roots inhibits primary root growth^4,5 and triggers an increase in root branching near the root apex.⁴ When the L-Glu was applied locally to the primary root tip at a concentration of just 50 µM, the inhibitory effect was as strong as when the same L-Glu concentration was applied to the whole root system,⁴ indicating that the root tip itself is the site of L-Glu sensing. The subsequent increase in root branching had the effect of increasing the root density within the L-Glu-containing zone, leading to the suggestion that it could potentially be a way to enhance the efficiency of soil exploration.^4,6 L-Glu concentrations within organic N-rich patches are likely to often exceed those needed to elicit the observed response.⁶

Although mitotic activity in the primary root tip is an early target of L-Glu inhibition, the mitotic divisions that are essential for lateral root initiation and subsequent development are evidently insensitive to L-Glu.⁴ This insensitivity, and the later onset of L-Glu sensitivity in the mature lateral roots (when they are >5 mm in length) are intriguing subtleties of the L-Glu effect that are crucial to the overall root architectural response.

L-GLu Appears to be Acting as an Exogenous Signal

As discussed in detail before, a number of the features of the L-Glu effect strongly suggest that the amino acid is acting as a signal rather than affecting root growth through changes in plant metabolism.⁴ Briefly, the most compelling observations that argue against a metabolic phenomenon are: (a) when a high L-Glu concentration (20-fold higher than the 50 µM needed for 80% inhibition of root growth) was applied to the majority of the root system, excluding the primary root tip, the growth of the primary root was not affected; (b) although glutamine (Gln) and Glu are rapidly interchangeable through the GOGAT cycle within the plant cell, even mM concentrations of Gln had no significant effect on root growth; (c) the L-aminobutyric acid or D-Glu) having no effect on growth.

Are Plant Homologs of Mammalian Ionotropic Glutamate Receptors Responsible for Mediating the L-GLu Effect?

There is evidence that plants possess Glu-activated ion channels with properties similar to those of the mammalian ionotropic Glu receptors (iGluRs) that mediate synaptic Glu signaling in the central nervous system (e.g., refs. ^7–9). Arabidopsis has a family of 20 AtGLR genes that are related to mammalian iGluRs,^10–12 and there is emerging evidence supporting a relationship between the AtGLR gene products and the Glu-activated channels.^13,14 A role for a GLR gene in root meristem function has been suggested by the recent isolation of a GLR3.1-defective rice mutant in which the activity of the root apical meristem is disrupted.¹⁵

As one approach to investigate whether iGluR-related receptors in Arabidopsis could be involved in mediating the effect of external L-Glu on root growth, we have examined the effect of a variety of agonists and antagonists of mammalian iGluRs. Such pharmacological tools have been frequently been used in the past to obtain evidence for iGluR-like activity in plants.^{5,8,12,16–20}

β-methylamino-L-alanine (BMAA) is an agonist of mammalian iGluRs that has been reported to stimulate hypocotyl elongation and to inhibit cotyledon opening and root growth.²⁰ As shown in Figure 1, we were able to confirm that BMAA inhibits Arabidopsis root growth. However, when we compared the effect of BMAA on two ecotypes that exhibit very different sensitivities to 50 µM L-Glu (C24 which is hypersensitive and RLD1 which is insensitive), we found no clear difference in their BMAA sensitivities and no correlation with their L-Glu sensitivities. When BMAA was applied at a concentration high enough to inhibit primary root growth in C24 to the same degree as L-Glu, the plants showed symptoms of toxicity and died (data not shown). We interpret these results as indicating that BMAA is not acting as an agonist of whatever receptors are involved in L-Glu signaling and that it is inhibiting root growth by a different mechanism.

Figure 1.

Effect of the iGluR agonist BMAA on root growth in two Arabidopsis ecotypes with contrasting L-Glu sensitivities. Seed of (A) ecotype C24 and (B) ecotype RLD1 were germinated and grown on vertical agar plates as described.⁴ Treatments were initiated by transferring four-day-old seedlings to agar plates containing BMAA or L-Glu at the indicated concentrations and primary root growth over the following 6 d was measured. The N source in all cases was 0.5 mM Gln.

We also tested three iGluR antagonists, 6,7-dinitroquinoxaline- 2,3-dione (DNQX), dizocilpine maleate (MK-801) and 2-amino-5-phosphonopentanoate (AP-5), and found that none affected primary root growth when applied on their own and nor were they able to antagonise the inhibitory effect of 50 µM L-Glu (Fig. 2). Our data contrast with those of Sivaguru and colleagues who found that 100 µM AP-5 was able to prevent inhibition of Arabidopsis root growth by 5 mM L-Glu.⁵ Apart from the much higher concentration of L-Glu, there are some significant differences between the phenomenon they were observing and the inhibitory effect described by Walch-Liu et al that might account for this apparent discrepancy.⁴ Most notably, the effect observed by Sivaguru et al was very rapid (minutes, rather than 24 h) and the primary target was the onset of cell elongation rather than mitotic activity.^4,5

Figure 2.

Effect of three iGluR antagonists on Arabidopsis root growth in the presence and absence of L-Glu. The experiments were performed with ecotype C24 as described in Figure 1. L-Glu was applied at 50 µM. (A) The concentrations of DNQX and MK-801 were 1 mM and 200 µM, respectively. (B) The AP-5 concentration was 200 µM.

Based on the experiments described here we were therefore unable to confirm that an iGluR-like receptor is involved in the observed root response to L-Glu. On the other hand, we cannot rule out the possibility that iGluR-like receptor(s) do mediate the L-Glu effect, but that they are of a type that is insensitive to the agonists and antagonists that are currently available. There is considerable divergence in sequence between plant GLRs and their mammalian homologs,^8,10,21 and we note that DNQX was also ineffective at blocking Glu-elicited changes in cytosolic [Ca²⁺] in Arabidopsis roots, even though it was effective in cotyledons.⁸

Concluding Remarks

It seems that insights into the mechanism of L-Glu sensing in root tips are more likely to come from molecular genetic approaches than from pharmacological ones. T-DNA or transposon insertion mutants are available for almost all the AtGLR genes (http://signal.salk.edu/cgi-bin/tdnaexpress), although functional redundancy and compensation between members of this large gene family may mean that techniques that can simultaneously target multiple genes will be needed (e.g., RNAi). We have also identified a number of L-Glu insensitive Arabidopsis mutants from a fast neutron bombardment population and are currently mapping the mutations responsible using genome tiling arrays. In a third approach, we have used populations of Col-0 × C24 and Col-0 × Ler recombinant inbred lines and near isogenic lines (NILs) to map quantitative trait loci controlling L-Glu sensitivity. Current data indicate a remarkably high level of complexity to the L-Glu effect, with L-Glu sensitivity being controlled by multiple genes and showing strong interactions with a number of environmental factors. The L-Glu insensitive mutants, and NILs differing in their L-Glu sensitivity, will also make valuable experimental tools for investigating whether, as hypothesized,⁶ there are particular soil conditions under which L-Glu sensitivity has a positive effect on plant fitness.

Abbreviations

AP-5

2-amino-5-phosphonopentanoate

BMAA

β-methylamino-L-alanine

DNQX

6,7-dinitroquinoxaline-2,3-dione

iGluR

ionic glutamate receptor

NIL

Near Isogenic Line

Pi

inorganic phosphate

QTL

Quantitative Trait Locus

Addendum to: Walch-Liu P, Liu LH, Remans T, Tester M, Forde BG. Evidence that L-Glutamate Can Act as an Exogenous Signal to Modulate Root Growth and Branching in Arabidopsis thaliana. Plant Cell Physiol. 2006;47:1045–1057. doi: 10.1093/pcp/pcj075.